In (highly contingent!) defense of interpretability-in-the-loop ML training
Published on February 6, 2026 4:32 PM GMTLet’s call “interpretability-in-the-loop training” the idea of running…
Category: Opinion
Spectral Signatures of Gradual Disempowerment
Published on February 6, 2026 3:08 PM GMTTL;DRAI disempowerment operates across markets, networks, and governance…
Demands Are All You Need: Prompt Imperativeness Drastically Reduces Hedging In LLMs (n=900, Cohen’s d = 2.67)
Published on February 6, 2026 1:22 PM GMTForewordIf you've ever noticed LLMs hedging considerably more…
If all humans were turned into high-fidelity mind uploads tomorrow, would we be self-sustaining?
Published on February 6, 2026 8:35 AM GMTThat is, would we in some sense manage…
AI benchmarking has a Y-axis problem
Published on February 6, 2026 7:45 AM GMTTLDR: People plot benchmark scores over time and…
Claude Opus 4.6 is Driven
Published on February 6, 2026 4:15 AM GMTClaude is driven to achieve it's goals, possessed…
Agent Economics: a BOTEC on feasibility
Published on February 5, 2026 8:15 PM GMTSummary: I built a simple back-of-the-envelope model of…
Moltbook as a setting to analyze Power Seeking behaviour
Published on February 5, 2026 8:07 PM GMTWe tested whether power seeking agents have disproportionate…
The nature of LLM algorithmic progress
Published on February 5, 2026 7:17 PM GMTThere’s a lot of talk about “algorithmic progress”…
Biotech Startup Stats
Published on February 5, 2026 6:40 PM GMTOne thing Claude Code has really made too…